LIDAR is the optical equivalent of RADAR. A LIDAR facility has been established at Rhodes University using a flashlamp-pumped dye laser as the transmitter and a photomultiplier tube at the focus of a searchlight mirror as the receiver. The setting up of the receiver and transmitter as well as the design and construction of the photon counting electronics is described. The LIDAR has been used to measure aerosol scattering ratios and temperature profiles in the stratosphere and these results are presented with the algorithms and software used to reduce the data. Finally some recommendations are made for future work

Durability is the most important factor in the design of heat-exchangers to meet the specifications of the customers. To predict the durability, endurance tests are carried out. In this thesis, one of the endurance tests, thermal cycling is performed for three different internal coolant flows until failure which is known as Wöhler test. A tool is developed in MATLAB that could import data from all kinds of endurance tests and visualizes time histories of every channel of the test (test parameters such as temperature, pressure, flow). An algorithm is developed for Level Crossing counting method which works based on the Markov cycle counting method. This produces Level crossing curves (LCC) for all the channels of the test which says how well a test has been performed and the total number of cycles of the test. It was observed that the LCCs obtained from the tool gives accurate results when compared with those obtained from LMS software whose approach of producing LCCs is Rainflow cycle counting method. Strain measurements are performed for the same flowrates as that of the Wöhler tests. The results from strain measurements and Wöhler tests are used in the determination of Basquin’s coefficient of the Wöhler curve. It was observed that the optimal value of Basquin’s coefficient is 3.4.

Let G be a finite permutation group acting on a finite set Ω. Then we denote by σk(G,Ω) the number of G-orbits on the set Ωk, consisting of all k-subsets of Ω. In this thesis we develop methods for calculating the values for σk(G,Ω) and produce formulae for the cases that G is a doubly-transitive simple rank one Lie type group. That is G ∼ = PSL(2,q),Sz(q),PSU(3,q) or R(q). We also give reduced functions for the calculation of the number of orbits of these groups when k = 3 and go on to consider the numbers of orbits, when G is a finite abelian group in its regular representation. We then consider orbit lengths and examine groups with “large” G-orbits on subsetsof size 3.

In this thesis, we propose to analyze panel count data using a spline-based sieve generalized estimating equation method with a semiparametric proportional mean model E(N(t)|Z) = Λ0(t) eβT0Z. The natural log of the baseline mean function, logΛ0(t), is approximated by a monotone cubic B-spline function. The estimates of regression parameters and spline coefficients are the roots of the spline based sieve generalized estimating equations (sieve GEE). The proposed method avoids assumingany parametric structure of the baseline mean function and the underlying counting process. Selection of an appropriate covariance matrix that represents the true correlation between the cumulative counts improves estimating efficiency. In addition to the parameters existing in the proportional mean function, the estimation that accounts for the over-dispersion and autocorrelation involves an extra nuisance parameter σ2, which could be estimated using a method of moment proposed by Zeger (1988). The parameters in the mean function are then estimated by solving the pseudo generalized estimating equation with σ2 replaced by its estimate, σ2n. We show that the estimate of (β0,Λ0) based on this two-stage approach is still consistent and could converge at the optimal convergence rate in the nonparametric/semiparametric regression setting. The asymptotic normality of the estimate of β0 is also established. We further propose a spline-based projection variance estimating method and show its consistency. Simulation studies are conducted to investigate finite sample performance of the sieve semiparametric GEE estimates, as well as different variance estimating methods with different sample sizes. The covariance matrix that accounts for the overdispersion generally increases estimating efficiency when overdispersion is present in the data. Finally, the proposed method with different covariance matrices is applied to a real data from a bladder tumor clinical trial.

La technologie « Radio Frequency Identification (RFID) » est devenue de plus en plus répandue dans le déploiement de diverses applications, telles que le contrôle des stocks et la gestion de la chaîne d'approvisionnement. Dans cette thèse, nous présentons une recherche systématique sur les problèmes de recherche liés au comptage et à la surveillance d’étiquettes RFID, deux composants fondamentaux dans les systèmes RFID, en particulier dans des systèmes à grande échelle. Ces problèmes sont simples à formuler et intuitivement compréhensible, tandis que tous les deux présentent des défis importants à la fois fondamentaux et pratiques, et exigent des efforts non négligeables à résoudre. Plus précisément, nous abordons les problèmes suivants allant de la modélisation et de l'analyse théorique, à la conception et l’optimisation de l'algorithme pratique. Pour poser les bases théoriques pour la conception et l'optimisation des algorithmes de comptage et la surveillance d'étiquette, nous commençons par étudier la stabilité de FSA. Techniquement, nous modélisons le “backlog’’ du système comme une chaîne de Markov, dont ses états sont la taille du “backlog’’ au début de chaque trame. Nous établissons ensuite un cadre générique d'estimation du nombre d’étiquettes RFID basée sur le filtre de Kalman pour des systèmes RFID statiques et dynamiques. Nous procédons en suite à résoudre le problème de détection des étiquettes manquantes, l'une des applications les plus importantes dans les systèmes RFID. Différent des travaux existants dans ce domaine, nous nous concentrons sur deux scénarios inexplorés mais fondamentalement importants, celui en présence d'étiquettes inattendues et celui des systèmes RFID multiple-région multi-groupe. Dans le premier scénario, nous développons un protocole à deux phases à base de filtre de Bloom. Dans le deuxième scénario, nous développons trois protocoles de détection d'étiquette manquante en intégrant une version améliorée de la conception du filtre Bloom. / Radio Frequency Identification (RFID) technology has been deployed in various applications, such as inventory control and supply chain management. In this thesis, we present a systematic research on a number of research problems related to tag counting and monitoring, one of the most fundamental component in RFID systems, particularly when the system scales. These problems are simple to state and intuitively understandable, while of both fundamental and practical importance, and require non-trivial efforts to solve. Specifically, we address the following problems ranging from theoretical modeling and analysis, to practical algorithm design and optimization. To lay the theoretical foundations for the algorithm design and optimization, we start by studying the stability of frame slotted Aloha. We model system backlog as a Markov chain. The main objective is translated to analyze the ergodicity of the Markov chain. We then establish a framework of stable and accurate tag population estimation schemes based on Kalman filter for both static and dynamic RFID systems. We further proceed to addressing the problem of missing tag detection, one of the most important RFID applications. Different from existing works in this field, we focus on two unexplored while fundamentally important scenarios, missing tag detection in the presence of unexpected tags and in multiple-group multiple-region RFID systems. In the first scenario, we develop a Bloom filter-based protocol. In the second scenario, we develop three protocols by incorporating an improved version of the Bloom filter design.

Diarrhoeal disease is a major cause of infant mortality in this and other developing countries. The assessment of the degree of dehydration in these children is often based on subjective findings alone. These have been shown to be inaccurate as an assessment of the degree of dehydration. Whole-body impedance (WBI) is a method of measuring total body water which is used to assess body composition. This WBI technology, which operates at a single frequency, has been applied to the assessment of dehydration in children with diarrhoeal disease. The normal range of WBI was determined on a group of normally hydrated children and was found to have a mean of 746 Ω, a standard deviation of 85 Ω and 95 % confidence interval for the mean from 720 Ω to 772 Ω. WBI did not depend on age, mass, height or sex. The WBI of a group of children dehydrated from infantile gastroenteritis was measured both before and after rehydration. The dehydrated group had a mean WBI on admission of 1089 Ω and a standard deviation of 149 Ω with a 95 % confidence interval for the mean from 950 Ω to 1109 Ω. This value was significantly greater than that of the normal group. The WBI of the dehydrated group after rehydration was not significantly different from the normal group. In addition a variable frequency bioimpedance analyser was designed and tested. It was found to have an error of less than 1 % over the frequency range 1 to 100 kHz. The device was evaluated on 11 normal and dehydrated children. Four dehydrated children were tested before and after rehydration and this data was compared to a standard electrical model for WBI. It was found that the model could represent the measured data over this frequency range. The extracellular resistive element of the model was mainly responsible for the changes seen during rehydration, suggesting that dehydration in gastroenteritis is mainly due to fluid loss from the ECF compartment.

No description available.

Philosophiae Doctor - PhD / Few studies of shallow water marine nematodes have been conducted around South Africa, and none from the west coast. Here, I analyse the composition of nematode communities from six stations along a 3 km transect in Saldanha Bay during both summer and winter, in order to describe the communities present and to explore the effects of sediment composition and heavy and trace metal concentrations on community structure. In order to put the local data into a global context, these data are analysed together with some consolidated data from elsewhere and patterns of richness and composition (at the level of genus and species) examined The transect in Saldanha Bay extended from below a mussel raft at one end into the bay, and six cores (35.7mm diameter) were collected at each station. All nematodes were counted and 100 randomly identified from each core. A total of 136 nominal species, 117 genera and 36 families were identified from both summer and winter stations. Nematode abundance was highest at stations under the mussel raft, which were characterized by high mud content and high concentrations of trace and heavy metals: diversity was comparatively low and the assemblage was dominated by a few, non-selective deposit feeders (especially Sabatieria). Abundance decreased, but diversity increased, with an increase in distance from the mussel raft, which was coupled with an increase in the particle size of sediments and a significant reduction in metal concentrations. There were three dominant (Comesomatidae, Desmodoridae and Linhomoeidae, present in 96%, 85% and 83% of samples, respectively) and four subdominant families (Chromadoridae, Microlaimidae and Xyalidae, all in 79% of samples) that were largely responsible for determining the community structure across the bay. Multivariate analysis of the data using PRIMER indicated that copper was the single variable that best accounted for the structure of the communities (70.1%), and the best 2-variable combinations were copper and organic nitrogen (70.3%), followed by copper, organic nitrogen and mean grain size (69.7%). Abundance was higher at all stations in winter than summer, and the results of the PERMANOVA test on station and season indicated that the variation in between Station- Season accounted for 27% of the differences in community structure. Although these results should be treated with caution owing to limited temporal sampling, they are similar to those obtained elsewhere in the world and indicate that nematodes can be used to study anthropogenic impacts in a local context. Despite the fact that Saldanha Bay has been subjected to industrial activities for more than thirty years, estimates of species richness for Saldanha Bay were surprisingly high: S= 136; ICE = 150; CFE= 173. As too were estimates of generic richness (S= 117; ICE = 131; CFE= 149), which were the fourth most rich of those global sites compared from similar depths. Incorporation of these data into a global dataset revealed an absence of any clear latitudinal pattern in the distribution of richness (genera or families), and there was no obvious geographic structure to global communities, based on the available data. These results suggest that genera and families are poor proxies for species (at the evolutionary level, but not at the ecological level) and they support the idea that everything is everywhere. Comments on ways that nematode research can be advanced in South Africa are made.

Diploma thesis deals with Ambient intelligence issue. Represents its basic characteristic and demands for its realization. Describes actual stadium in this concept development and shows some present projects. Next focuses on Intelligent buildings issue. In connection with this addresses to so-called human behaviour patterns. Various methods of human behaviour patterns measurement are discussed there. Thesis then focuses on people counting system design, which is based on camera record. Such system represents way of human behaviour patterns measurement. Lastly, the using of this way obtained data is discussed.

Studiens syfte är att beskriva hur elever i de lägre åldrarna använder handgester när de löser aritmetikuppgifter. Den teoretiska utgångspunkten i studien är embodied cognition, där det centrala i teorin är att vi lär oss genom kroppen. Tio elever i sjuårsåldern i Sydafrika har observerats med fokus på deras användning av handgester. Resultatet visade att flera av eleverna använder handgester när de löser additions- och subtraktionsuppgifter och att det finns ett antal olika handgester de använder sig av. De olika handgesterna delades in i taktila och icke-taktila gester, som i sin tur kopplades till konkreta och abstrakta gester. Slutsatsen av studien är att elever använder sig av olika handgester i olika kombinationer när de löser aritmetikuppgifter. Deras gester med händer och fingrar kan vara ett uttryckssätt för deras utvecklingsprocess och kan fungera som en bro mellan konkret och abstrakt tänkande. / The aim of the study is to describe how students of the lower ages use hand gestures when solving arithmetic tasks. The theoretical basis of the study is embodied cognition. The viewpoint of embodied cognition holds that the body is a tool for learning. Ten students in South Africa, aged seven, have been observed based on their use of hand gestures. The result showed that several students use hand gestures when solving addition and subtraction tasks and there were a number of different hand gestures being used. The various hand gestures were divided into tactile and non-tactile gestures. These could in turn be linked to concrete and abstract gestures. The conclusion of the study is that students use different types of hand gestures and also in various combinations when solving arithmetic tasks. Their hand gestures can be a way of expressing their development process and to function as a bridge between concrete and abstract thinking.